The economic consequences of religion

Barcelona GSE Focus

Religiosity has gained a central stage in today’s societies. From terrorism to political debates concerning same-sex marriages, religion has emerged has a key component of modern societies, even secular ones. While the sociological aspects have been and are being thoroughly studied, it is not clear whether religion can have an impact on economic outcomes and if yes, how.

At least since Weber’s contribution, the different kinds of culture that religions can create have been seen as potential contributors to differences in economic development. Nevertheless, a clear theoretical channel has not been proposed yet. Understanding how religiosity can impact economic outcomes is key to understand the channels through which religions can influence economic growth or even redistribution and inequality in modern societies.

Joan-Maria Esteban, Gilat Levy and Laura Mayoral in their Barcelona GSE Working Paper (No. 843), “Liberty, Religiosity and Effort,” shed some light on these issues. First, they provide a theoretical channel through which religiosity can have an impact on individual preferences and, thus, on labor supply. In particular, they propose a theoretical framework that allows for the study of the joint effect of religiosity and civil liberties on labor supply. Second, they empirically test the theory with data from European countries and, in doing this, they create a new index of civil liberties.

By the sweat of your brow you will eat your food…

The authors propose a simple theoretical model that fits perfectly the situation under analysis. Individuals in their stylized society have to decide, as it is standard in economics, between how much to consume and how much effort to put into work. Obviously, the more effort is put into labor, the more an individual is able to consume. Nevertheless, individuals face the standard trade-off between a desirable option, i.e. consumption, and an undesirable one, i.e. labor effort. The novelty of the proposed framework is that it introduces two new variables. The first one is the individual level of religiosity. The second one is the presence of civil liberties in the society, e.g. possibility to divorce, LGBT rights, abortion, etc. The mere presence of such liberties in a society, independently of individual usage of them, can have an impact on individual welfare as current political debates on the subject highlight.

The key assumptions the authors make is that the stronger is the degree of religiosity the stronger is the distaste for these liberties. Notice that this implies that, ceteris paribus, a higher presence of liberties in the society, decreases the evaluation of consumption and leisure for religious individuals. This assumption is in line with the idea that religions determine the behaviors a pious society should comply with. Thus, the more religious is an individual and the more her society is far from the ideal structure her beliefs imply, the less she enjoys living, i.e. consuming goods and free time, in such a society. Obviously, such effect of liberties on individual welfare may very well depend on the kind of religious affiliation an individual has, hence the authors allow for different individual sensibilities depending on the different religious affiliations, secularism included.

The mechanism modeled leads to two important and novel predictions. As long as there is a larger complementarity between liberties and consumption than between liberties and effort, (i) an increase in religiosity decreases labor supply and (ii) an increase of civil liberties increases the effort of secular individuals while decreasing the effort of religious ones. These results are due to two main reasons.

The first prediction is a consequence of the fact that an increase in religiosity, by enhancing the negative effect of civil liberties on individual welfare, has two byproducts for any given level of civil liberties. First, it decreases the evaluation of a positive change in consumption and, second, it decreases the evaluation of an increase in leisure, hence increasing the evaluation of effort. If liberties have a stronger complementarity with consumption than with leisure and, thus, effort, the first effect dominates the second, therefore more religious individuals will be more willing to decrease consumption by working less.

Similarly, the second result is due to the fact that religious and secular individuals are affected in opposite ways by the presence of civil liberties in their societies. Religious individuals dislike the presence of civil liberties and this negatively affects their valuation of consumption and leisure. Hence, a higher level of civil liberties implies a lower valuation of changes in consumption and leisure. The conclusions are opposite for seculars. Given that civil liberties have the positive effect on them, the higher the liberties the higher the valuation of an increase in consumption and leisure. Hence, an increase in civil liberties, given the assumptions on complementarities between consumption, leisure and liberties, pushes religious individuals to substitute consumption with leisure, while the opposite happens for seculars.

The authors are able to propose a clear channel through which religions can have important economic effects. Will these theoretical predictions survive the trial of fire? Are the results empirically valid?

Stop doubting and believe

To empirically test their theory, the authors first construct a new index of civil liberties for different European countries that reflects the legal evolution of civil liberties from 1960 to 2013. They build indices that track the evolution of the legal status of each issue of interest, e.g. one index for abortion, another one for divorce, etc., and then they create an aggregate index that is the average of all individual indices.

After having created the index, the authors use individual-level data from the European Social Surveys to explore empirically their theoretical predictions. The results are statistically significant and in line with the theory. In particular, an increase in one standard deviation in the intensity of religious beliefs is associated with a decrease of 0.34 weekly hours of work on average for a median level of liberties and a decrease of 2 weekly hours of work for a high level of liberties.

These results are robust to many different specifications and take into account the possible endogeneity of individual religiosity. That is, religiosity might be related with unobservable individual characteristics that influence labor supply. To avoid this problem, the authors construct an instrument for individual religiosity based on the assumption that religion is a cultural trait that transcends national borders. To measure the religiosity of a particular individual in a given country, the authors use the average of individual religiosity of socioeconomically similar individuals in neighboring countries. As long as religious affiliation can be considered exogenous, this way of instrumenting the variable of interest assures that all possible interactions between labor supply and individual unobservable characteristics through religiosity are excluded.

The empirical results confirm that religions can have relevant economic effects that have to be taken into account when considering the overall consequences of religions and civil rights on the development of the complex sociological construct modern nations represent. In particular, the results presented in the paper imply that more secular societies should enjoy higher levels of GDP given the positive effect of liberties on labor supply in those societies, but they should also experience higher degrees of inequality the higher the level of civil rights due to the diametrically opposite effect of liberties on labor supply and thus labor income of secular and religious people. These effects cannot be overlooked for the construction of a prosperous society.